Top-Bottom 3D vs. Side-by-Side on Passive Display - Page 2

If active has that effect on you, but passive doesn't, then get a passive set. It doesn't matter that it's only 540 lines per eye, at least it's 3D.

I doubt however that passive will help you, as I think 3D in general is adverse to you. Jerking back and wincing is an extreme reaction and I doubt it's caused by the glasses format.

AFAIK Passive 3D TVs work almost the same as the interlaced CRT TVs. With passive first a 1920x540 a left eye odd line image is displayed followed by a left eye even line image then the same is done for the right eye. This sllows the mind to integrate the even and odd line images into a full 1080p image for each eye.

Please read this carefully and reconsider your opinion about passive 3D being 1080p for each eye.

http://www.hometheater.com/content/closer-look-active-vs-passive-3d-flat-panels

also this video may help your understanding.

http://www.youtube.com/watch?v=ETc3Ep3wcEk

The argument for passive being 1080p isn't based on the brain fusing the simultaneously-displayed 540 left-eye lines with the 540 right-eye lines like that video suggests. It's that the brain fuses the 540 lines shown for each eye at 1/120 sec with the the 540 lines shown after the screen refreshes at 1/60 sec. Since you can't physically measure the resolution formed inside the brain, the "540 + 540 = 1080" theory can't really be tested quantitatively.

That said, active quite possibly has a higher effective resolution though since it fuses 3D from alternating 1080p images rather than fusing 1080p from a refreshing 3D image.

This is addressed to BleedOrange11. The above quote is from a passive fan. He sees slight aliasing on his passive display. 1080P shows no aliasing viewed from any distance. This is an unbiased real life observation from a person that definitely favors passive. Can we put to rest the idea that passive displays 1080p in the brain or in any other way.
Edited by Robut - 7/12/12 at 2:50pm

The aliasing might not necessarily be the result of a drop in resolution, although it could be (I honestly haven't given it much thought). My assumption is that, because of the line separation from the interleaved image, there's no "blending" happening on those pixels because they're not actually right next to each other (although the eyes perceive that they are), hence the aliasing. It's like turning AA off in a videogame.. it's still 1080p, just jaggy. And I can tell the difference between a 1080p game with AA on and AA off in 3D, for whatever that's worth. And I'm more susceptible to seeing aliasing because of my work in visual effects.. I can't leave any kind of jagged edge in a final render, regardless of resolution, so I tend to be able to spot it very quickly (that's why I said I'm "trained to look" for aliasing). Take the "Big Buck Bunny" animation that everyone likes to use in reviews.. there's plenty of aliasing in that one, probably because the creators didn't have the means or the time to completely eliminate it. And I'd wager 95% of the people watching the video never even notice it's there. I'm not arguing that it's not there on a passive display, it definitely is, it just may not be enough to bother some people. I can see it, and I'm still okay with it, so take that as you will.

BleedOrange11, where are you getting the idea that the images are being displayed sequentially? That's active-shutter technology.. with a passive display, the two images are shown simultaneously, there is no "switch" from a 540 left-eye image to a 540 right-eye image, and it's not done in an "interlaced" fashion.. they're shown at the same time, progressively. I think you're confusing "interlaced" and "interleaved".. they're not the same thing:

With an interlaced display, let's say an old CRT TV, it would display left-nothing-left-nothing-left-nothing-left-nothing, all the way down, then go back to the top and display nothing-right-nothing-right-nothing-right-nothing-right, all the way down, coming in after the left image was displayed, and filling in the other half of the image. That seems to be what you're describing, but modern LCD monitors don't work that way (to my knowledge, they never have).

With an interleaved image (as my monitor supports), both eyes are drawn at the same time, top to bottom. It's still a progressive display like any other LCD.. it's left-right-left-right-left-right-left-right-left-right-left-right-left-right, all the way down, and then it moves onto the next frame, not the next eye. Checkerboard displays work the same way, both eyes are displayed at the exact same moment in time, not sequentially.

The output of those two may look very similar to the naked eye, but the methodology behind them is very different.

Also, the 120Hz argument falls apart from the simple fact that mine is an IPS display.. it's capped at 60Hz. 120Hz is only a requirement on active-shutter displays because of the way that technology works. Although it can be found on TN-panel passive displays, it's not actually required for the 3D technology to work. In fact, now that I think about it, I suppose the only difference between my monitor and any other IPS monitor is the polarization/retarder layer added to the panel, and the firmware that allows it to decode a native 3D image (like output from a standalone player over HDMI) into interleaved 3D (which I'm not even using, I do all of my 3D through software). When 3D is turned off (as it is most of the time), it's just another monitor.

For the record, I'm not trying to argue that passive is "better".. as others have stated, each one has pluses and minuses, and a lot of it comes down to personal choice. But I did do quite a bit of research on all kinds of 3D technology (I'm a computer tech by trade, so I like to know exactly how my technology works), and it was my decision to go passive. Also bear in mind this is just a computer monitor and not a TV.. I may decide differently when it comes time to buy my next big-screen.

I would think that aliasing around high contrast areas should be expected with passive and active 3D LCD displays due to their crosstalk issues caused by slower pixel response times. A tiny bit of crosstalk will be perceived as blurriness rather than obvious ghosting.

And no, his one observation is not conclusive evidence that the brain cannot fuse lines to create 1080p.
Edited by BleedOrange11 - 7/12/12 at 5:40pm

Three facts that completely contradict your logic:

1. Resolution is not measured in the brain, it is measured with test patterns on the display. If 540 lines are displayed, resolution is 540, period.

2. If your argument that the brain combines two 540 images to make one 1080 were true, then it would also have to be true that it combines two 1080 images to make one 2160 image, which means that passive is still half resolution, even to the brain.

3. Now if the so called first pass at 1/120 were shown on the odd lines and a second pass of lines at 1/60 second were displayed on the even lines, you might have had a case, except that the polarized filter prevents this from being possible. All images for one eye must be on the even lines and all for the opposite eye must be on the odd lines, as there are oppositely polarized filters physically positioned in front of each line.

Again, no argument about which is better, I could care less what you watch and think it's great that people have choices.

The aliasing is caused by the stair stepping of diagonal lines, as every other line is missing from the image, and the image "jumps" to the next line.

_____
______
_______

vs.
_____

______

I think you may have misunderstood my previous post. I never said anything about passive being interlaced or displaying images sequentially, and I don't disagree with how you've described the tech.

My understanding of passive 3D 120 Hz LCDs is that at any one time, odd lines show 540 pixels of a right eye image and even lines show 540 pixels of a right eye image. Then after 1/120 sec, the display refreshes and the other 540 pixels for the right and left eye images are displayed. After 1/60 sec, the brain perceives two full images (one for each eye) by fusing the lines together.
Edited by BleedOrange11 - 7/12/12 at 6:04pm

1. After 1/60 sec, 1080 lines have been displayed to each eye for both passive FPR and active displays.
At 1/120 sec: Passive shows 1/2 of a left-eye and 1/2 of a right-eye image. Active shows one full left-eye image, and no right-eye image.
At 2/120 sec: Passive shows other 1/2 of left- and right-eye images. Active shows one full right-eye image, and no left-eye image.

2. No, the argument is that passive combines four 540 half-images to make two 1080 images.

3. One of us must be confused about how passive works, and it very well could be me. How could you possibly perceive a full image for each eye that's not broken up by lines if the right- and left-eye images do not alternate lines though?
Edited by BleedOrange11 - 7/12/12 at 6:29pm

For those who may not understand how it works:

While normally light scatters in every direction, light bouncing off a flat surface, such as water, becomes predominantly polarized by the direct reflection. Thats how polarized sunglasses reduce glare. The lenses block light more in one direction than the other, which means they block most of the reflected light while allowing other light through.

To visualize it, consider the the waves of reflected light, or glare move mostly in this direction ||||||||||||||. The filter of the glasses is arranged at a 90 degree angle or
======= so they will tend to allow only horizontal waves through. One can test this themselves at the gas pump. While reading the LCD on the pump (while wearing polarized sunglasses) rotate your head sideways, and notice how the numbers become harder or easier to read. This is polarization at work. On a 3D TV, the lenses of the glasses are oriented 90 degrees from each other, in order to make 3D work.

On a passive TV screen, filters one line high are placed in opposite arrangement for each line, every other line being left oriented, and their counterpart being right oriented



This dims the picture slightly (as any filter would), except when the light goes through another lens on the glasses which is oriented one of the two ways,

........... L.................... R
//////////////////// \\\\\\\\\\\\\\\\\\\\
//////////////////// \\\\\\\\\\\\\\\\\\\\
//////////////////// \\\\\\\\\\\\\\\\\\\\



the eye sees this.



or this,



depending on which way each lens is polarized, because the light is blocked completely (in theory) by the opposite polarization, but passed through the similarly polarized lens.

Both the left and right eye views on the 3D disc are stripped of every other line (for example all the even lines from the left eye image and all the odd lines from the other image), and the two can then be interleaved together into one frame. The filters on the screen, as shown above, in combination withthe polarization filters in the glasses allow only the appropriate lines to be seen by each eye.
Edited by Augerhandle - 7/12/12 at 7:01pm

So why doesn't the brain perceive two images with lines spaced in between them? Does the brain just make up info to fill in the gaps like it fills in the eye's blind spot (where the optic disc is located)?
Edited by BleedOrange11 - 7/12/12 at 6:42pm

See my post above. What you describe would superimpose the second pass of 540 lines over the first, confusing and blurring the image. The blank (black) lines would still be between the image lines, and only 540 (now blurred) lines would still be shown to each eye. The eye would combine them, but not as 1080 separate lines. They would appear as 540 confused lines.

For it to display all 1080 lines, the filters on the screen would have to shift up or down one line, which would make the screen active. Making the screen active would still allow cheap glasses, but would make the TV prices prohibitively high. Another solution would be to alternate the polarization of the glasses every 1/120 of a second. In either case the system would no longer be passive, would it?

Kind of. What the eye perceives is the aliasing that people talk about. Consider looking through a chain link fence or a screen door. Depending on how one focuses or how far away one is, the grid can seem to disappear.

The same thing occurs with "screen door effect" with plasmas and LCDs. the brain ignores the 30% of the screen made up of the black grid between pixels if one sits far enough away, and the grid pattern gives the picture a sense of what some people interpret as improved sharpness.

Okay, well thanks for explaining that. I must have read something earlier that had me confused.

YW

Wait a minute. What's this about? (I'm still reading the rest of the article.) http://www.hometheater.com/content/closer-look-active-vs-passive-3d-flat-panels

EDIT
Yeah, I think I was right earlier. This article better explains how passive works. I'm going to stand by my original argument of passive creating "fused 1080p." (Not arguing that it's free of temporal compression artifacts or that it has an effective resolution equal to active 3D--just that you see all 1920x1080 pixels for each eye after 1/60 sec on FPR passive 120+ Hz models.)
http://www.hometheater.com/content/passive-3d-resolution-update

I guess the brain somehow merges each eye's odd and even lines correctly even though they're showing up in the same location on the TV.
Edited by BleedOrange11 - 7/12/12 at 9:48pm

He just described what you did, odd and even being diplayed on the same line 1/120 of a second apart. They would be combined to form 540 blurry lines.

From the same article:

If LG is doing this, it is to be able to claim that "all 1080 lines are being displayed". It doesn't mean that the image is being shown correctly, or that it is 1080 per eye, as explained above.

Didn't see this article posted in this thread yet, although I know it's been debated elsewhere.
still, much more detail than other links in this thread IMHO. (and displaymate is well known for years) granted it's from 2011 but still an interesting read I think and has some surprising test data/notes. (like the small text test)

http://www.displaymate.com/3D_TV_ShootOut_1.htm

My son in law recently bought a LG 55LM7600 - and honestly we are all impressed at the 3D. Easy on the eyes, literally no crosstalk, ghosting, etc . Very impressive to me and I was never a fan of 3D before. I've only seen active in stores/demo rooms for short periods, but all things considered I think his LG 7600's passive 3D is impressive. Light glasses, great viewing angles, not a big hit on brightness in 3D.
Its an IPS screen, so blacks not as deep as some other panels or plasma, but for an edge lit LED he must have gotten a 'good one', as no issues with flashlighting and or clouding, but I've read lots of posts from others not so lucky. (makes me wonder what the return rate is on edge lit TVs.)

anyway, the above article is interesting I think but I know many disagree with their results/conclusions...

The study you reference has some noticeable examples of bias. The displays tested were 47". At that size the artifacts exhibited by passive are less pronounced. The active displays tested were not top of the line models and did not include plasma displays which produce the best active 3d. It also excluded DLP and projection 3d. So to me the conclusions reached are not entirely valid as a comparison of active vs passive.
The study has been referenced dozens of times in this forum.

All I'm trying to argue about are the claims that passive is 1080p somehow. I think augerhandle has explained how this is not possible very well. That said, I think passive 3d is a good experience possibly better than active. It's just not 1080 resolution.
Edited by Robut - 7/12/12 at 8:41pm

Agreed. That study pretty much says passive FPR LCDs have better effective resolution in high contrast areas than early active LCDs due to less crosstalk issues. It can't really be used as evidence for any other claims regarding resolution and active vs. passive.
Edited by BleedOrange11 - 7/12/12 at 8:52pm

Yeah, I know that article has been posted/debated before, but as I said in my first post just didn't see it in this thread. and again it's from 2011, hotly contested but it does have some tests related to comments here. (I think this is another case where I regret posting/wasting all our time lol)

anyway, my SIL went thru 2 plasmas and returned them (bad luck I guess, buzzing, didn't like the battery glasses (has kids), saw some IR, etc) - then saw a review of the LG 6700 at televisioninfo (dated 3/21/2012) - http://www.televisioninfo.com/content/LG-47LM6700-3D-LED-LCD-HDTV-Review.htm that said in the summary - "The best 3D in the business", so went looking for one and got the 55LM7600 for $1650 (I think) w/12 pairs glasses, including some kids size. 3rd time was the charm for them. Not a perfect TV of course, but a great fit for them. And it's almost made a 3D convert out of me. Hugo 3D and Journey 2 looked great and some Tahiti 3D movie I remember as being very very sharp in 3D. zero complaints from any of them or me on the 3D image quality. (But I didn't care at all for the ("wii"-like) magic remote - no input button???. EU models had both remotes but US buyers don't - cost $13 or so for the std remote.)

it's a 55in with the couch appx 8ft away (just under, less than the 'recommended' I think) - no "artifacts" that any of us have seen nor are lines visible at that distance. (Yes, very close you can see lines in 3D, but who sits that close.) Not what I expected after reading all the comments about passive 3D rez.

But again I'm sure there's better out there (and already commented on IPS panel/black level, edge lit, etc earlier above) but he could have done a lot worse. All things considered, the LG 55in/7600 seems perfect for them. (No buzzing, IR, no headaches, no batteries in glasses, no flicker, crosstalk, very wide viewing angles, no issues seen with 24fps movies or fast motion 3D or 2D)

I always thought if I ever bought a 3D tv it would be an active/plasma, but not so sure now. But by the time I'm in the market things will have changed/improved again and I'd definitely look at active and plasma again. (knock on wood my CCFL LCD (2D) still going strong ... but I probably jinxed it by saying that)

Didn't want to stir the pot on passive vs active (again) but wanted to say how impressed we have been with the 2012 LG 7600's 3D. After seeing all the comments on passive 3D in forums for years, I would never have expected it to look as good as it does (IMHO).

BTW - They did see some 3D problems (severely out of sync on fast motion) with some specific 3D movies (most were 20th Cent. Fox titles - Ice age DOTD, RIO but also the 3 Musketeers) with a Sony BDP-S790 that had been updated w/firmware from May 2012 (nothing later yet). Tried them in a Yamaha A1010 and a BB store's Panny player - worked fine (same TV/settings). So returned the S790 for another one (with original/earlier firmware) and that one was fine, so he didn't upgrade the FW. There's some posts here and at bluray.com (S590 owner recently) about this issue. (and some older reports on the same thing with some other players, some fixed in fw updates already.) S590 owner reported it to sony so hopefully they'll fix it soon. Just a FYI as some may blame their TV.

I don't know how highly thought of that tvinfo site is but they have tested plasmas also (and rated them higher overall, esp. for 2D pix quality). Their reviews do seem better than some but that's just my opinion. Nothing more I can contribute here so I'll bow out. (Didn't mean to write a book.) Have a great weekend.
Edited by Xlr8yourMac - 7/13/12 at 12:04am

Maybe we should call passive FPR "1080x120i." Does that work for everyone?

I'm sorry but not me. A 1080i image doesn't have black lines that are visible at close viewing. To me 1080i is hard to distinguish from 1080p.

Why make up names, when there are industry standards already in place and arrived at years ago by expert scientists and engineers who work in the field?

It's 1920x540P per eye.

You like it, so watch it. Tell others to buy it. It doesn't matter to me. No need to pretend it's something it's not though. If it makes you feel better, when someone mentions that it's 540 just answer "so what?".

Ah, okay, I see what you're saying now. That's the new method that LG is working on. It seems that would fudge the image even more with the lines shifting up and down.

In any event, I'm perfectly happy with my monitor. It's not perfect, but there's no such thing as perfect 3D yet.

There is no industry standard for what LG's passive tech is called, and it shares traits of both interlacing and progressive. LG calls it 1080p, but it's not. To my eyes and at a viewing distance behind 5 feet, it looks much better than 540p and almost equivalent of 2D 1080p, but slightly softer. It's somewhere in between, which is why I've been calling it "fused 1080p" and stating "effective resolution not equal to active."

Sorry, I'll quit squabbling over technicalities after this post, but if you see all 1080 lines, and they get refreshed 60 times per second, then isn't it technically 1920x1080x60? I suppose it depends on the definition of 1080. (Do the lines have to show up in different locations on the TV?) The technique is neither progressive nor interlaced. It's something different--maybe interleaved?
Edited by BleedOrange11 - 7/13/12 at 9:13pm

There is an industry standard name for LG's passive tech, it's called Film Pattern Retarder. LG calls it Cinema 3D. I've already explained how it works. The picture is not interlaced. The left and right views are interleaved together, and then displayed progressively.

How can you say it looks much better than 520P, when the only 520P you have seen is the TV you're talking about? That's like looking at your own reflection and saying it looks much better than your own reflection. To your benefit though, with your 42" screen size and at that distance, you may not be able to discern much difference in resolution. It doesn't change the resolution however, just your impression of it. That has been covered here. http://s3.carltonbale.com/resolution_chart.html According to that article, you wouldn't be able to see any difference between 1080P and 480P (standard def) on your TV at a distance of 12 ft.

Just because it is half the lines doesn't mean that it has to appear half as good to the eye, either. Cut your speaker power in half, and the sound drops 3 dB. A typical listener can barely detect a change of 2 dB. So the volume will barely seem to change at half the power. It would not be correct to say that the power is "somewhere between 540 and 1080 watts" or "it sounds much louder than 540 watts". It is just 540 watts.

Also, you do not need to name what it is, as the experts have already done that. Instead of making up terms like "fused 1080P", just call it passive, then people will know what you mean. BTW, the term "effective resolution", is already being used as well (by those pesky experts again) in regards to scaling, so you shouldn't use that in this case either. You'll only confuse the subject more.
1080 lines, progressively displayed and refreshed at 60 frames per second equals 1920x1080P60. Refreshed at 24 times per second it is1920x1080P24. It does not "depend on the definition of 1080". The definition of 1080 does not change from one person to the next, it only gets misinterpreted, as not everyone is up on the technology side of things.
http://en.wikipedia.org/wiki/1080p

At least 1080 lines, not 540, not 720, but 1080.

Here's one more thing to think about.


It is recommended to sit closer while watching 3D than while watching 2D, and there is a way to calculate the best distance.

According to the calculations explained in this article: http://www.practical-home-theater-guide.com/3d-tv-viewing-distance.html

and using your TV as an example, a 42" diagonal measurement calculates to an approximate width of 36.6 inches. Using the formula above, the optimal viewing distance would be 1.4 x 36.6, or about 51.24 inches, which equals 4 ft 3-1/4 inches.

You stated that it looks better than 540 at more than 5 feet, and the math shows you why. If it was 1080, you should be able to watch at 4 ft with no degradation.

According to the same article, if you sit farther than 5-1/2 feet while watching 2D, you wouldn't be able to make out all the detail in 1080P. You might as well have a 720P TV or get a larger screen, because the resolution is wasted at farther viewing positions.

This explains why passive looks fine to people who watch on smaller TVs, as from what I've been reading, most people sit 10-12 feet away!

Just for reference, I typically watch 3D on my 65" active DLP set at around 6 ft., and watch 3D on my son's 73" from about 7-1/2 feet, which fits in with the formula. In comparison, passive at those screen sizes would be impossible to watch.